[Coco] EEPROM replacement for 27C256

[Darren A]

On Sun, Jun 28, 2015 at 11:55 AM, Gene Heskett  wrote:

> Another piece of trivia folks.
>
> As long as the data buss is the same width AND on the same series of
> pins, it doesn't matter one bit if data pin d0 is connected to the data
> line on the board named d0. Ditto for d1-d7 (or d15 for 16 bit wide
> chips) Because any sucn accidental or on purpose scrambling simplifies
> the board layout, scrambling of the data written, it will be unscrambled
> by the same "re-arrangement" on the subsequent read.  You will get back
> the same data you wrote in any event.
>
> Exactly the same situation exists for the address lines, the operative
> again is that the same group of pins are being used.  The a0 thru a15
> (or a23, a31 in the case of wider chips) can be fed to the chip in any
> sequence.
>
> As long as those conditions are met, it will work as expected.
>
> I can think of a flash eprom that would need exact matching because they
> often have to erase and rewrite a whole 4k block to change one bit.  The
> cocosdc might, probably is subject to that exception, but Darren can
> correct me.
>


Scrambling the data and/or address lines is great when dealing with RAM but
I never do that with EEPROM or Flash.  As you mention Gene, those devices
often utilize sector or paging schemes for erasure/programming, so you'd
probably want to have at least the address lines wired correctly.  There
are also Software Data Protection schemes to think about and whether or not
you want to write to the device outside of the circuit (use a chip
programmer).

- Darren